标题:Photographer trajectory detection from images
作者:Fan, Linwei; Li, Huiyu; Li, Mengjun; Zhang, Yan; Li, Jinjiang; Zhang, Caiming
作者机构:[Fan, Linwei; Li, Huiyu; Zhang, Caiming] Shandong Univ, Sch Comp Sci & Technol, Jinan 250101, Shandong, Peoples R China.; [Fan, Linwei; Li, Huiyu; Z 更多
通讯作者:Fan, LW;Fan, LW;Fan, Linwei
通讯作者地址:[Fan, LW]Shandong Univ, Sch Comp Sci & Technol, Jinan 250101, Shandong, Peoples R China;[Fan, LW]Coinnovat Ctr Shandong Coll & Univ Future Intelli, Ya 更多
来源:PERSONAL AND UBIQUITOUS COMPUTING
出版年:2018
卷:22
期:5-6
页码:1005-1015
DOI:10.1007/s00779-018-1150-5
关键词:Image matching; Image similarity; Feature point; Image context;; Essential matrix
摘要:This paper proposes a novel method for detecting a photographer's shooting trajectory based on select images. Firstly, in a Lab color space, directional information and perceived color information were combined, and similar images were found by a color difference histogram. Local invariant descriptors were then constructed by the contrast context histogram method to match feature point areas and their context, and to judge whether these areas corresponded. Through this, the corresponding relationship for feature points between image sequences was obtained. Furthermore, the essential matrix for a pair of images was obtained through the singular value decomposition method to determine photographer trajectories.
收录类别:EI;SCOPUS;SCIE
资源类型:期刊论文
原文链接:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85046465492&doi=10.1007%2fs00779-018-1150-5&partnerID=40&md5=fbd88ebe3d2deb7e3565513db72edfe5
TOP